Galvanic detection of gold on PCBs

Chat about pick and place machines, reflow ovens, assembly techniques and other SMT tips & trix

Moderators: adafruit_support_bill, adafruit

Please be positive and constructive with your questions and comments.
Locked
aleksolder
 
Posts: 5
Joined: Sat Nov 09, 2013 5:14 am

Galvanic detection of gold on PCBs

Post by aleksolder »

In taking some surface potential measurements of odds and ends, I noticed that the surface potential of some allegedly ENIG PCBs I have show values corresponding to brass (or thereabouts). My values for SS from the tableware, copper from wire, and lead from fishing tackle show sensible values. But I still suspect something wrong in my procedure or assumptions.

Since you folks are very meticulous, you probably know why I'm getting the brassy value where I expect a goldy value. Would you please explain?

My procedure: Mix table salt with tapwater to make brine, any proportion. Place two metals into a cup of this brine, not touching each other, and with some portion of each sticking out of the water. Connect the red test lead of a voltmeter to one metal. Connect the black to the other metal. The measured value is the difference is surface potential. Look up the measured value on a galvanic series and you can make a good guess about what the metals are. If you know one of them, you can see from the series what the other is, with high certainty. The metal must not have any oxidation or crud on it; use chemical or mechanical aid to clean, but if you do, immerse and wait for the value to settle (removing some of the surface will cause the surface potential to temporarily climb beyond the metal's natural value). Don't abrade ENIG: you will get copper!

User avatar
adafruit_support_mike
 
Posts: 67446
Joined: Thu Feb 11, 2010 2:51 pm

Re: Galvanic detection of gold on PCBs

Post by adafruit_support_mike »

The main problem is that you don't have a bar of pure gold. You have a gold flash over a sheet of copper foil.

There's an intermetallic boundary between the gold and the copper beneath it, and that will have a small voltage determined by the electronegativities of the two metals.

There's also current flowing out of the gold and into the liquid, and the dynamics there are complicated.

Copper is a better conductor than gold in general, and the copper foil is much thicker than the gold flash on top of it. From a point on the gold/liquid interface looking sidways along the surface, the resistance through the gold will be much higher than the resistance to the copper beneath the gold.

That means most of the current entering the liquid will come through the gold from the copper directly beneath it, not through the gold. That means you have a small voltage gradient based on the relative resistance between the two metals that overlaps the electronegative potential difference.

The combined voltage gradient will attract charged particles from the liquid, so the gold will act like the gap in a very leaky capacitor.

I honestly don't know how all those effects will add together, but it's pretty likely that the combined effect will be different from what you'd get at the surface of a bar of pure gold.

aleksolder
 
Posts: 5
Joined: Sat Nov 09, 2013 5:14 am

Re: Galvanic detection of gold on PCBs

Post by aleksolder »

Thanks very much for taking trouble to look at this. As it happens, I have recalled the answer:

I didn't says so, but the readings were very unstable, which corresponds with expectations for a highly noble metal, such as gold or titanium. My readings for the other materials were stable, which should have made me remember this point.

Nobility can be described as an unwillingness to participate in electrochemical reactions, so this technique, which relies upon reactions with the electrolyte, is not reliable for them.

Addressing some of your points: Everything happening below the surface is not relevant in this technique, since only the surface interacts with the electrolyte. The flow of current anywhere in the circuit is not relevant (potential is just potential, after all). Some simplification here, but good enough guidance for this procedure.

The conclusion is that you cannot use this technique to directly test for gold at the surface, though you can use it to test for almost any other metal, such as brass, tin, steel, lead, copper, etc, and you might be able to say with some confidence (I don't know how much) that with unstable readings, the surface metal is probably noble, and the color would tell you whether it's gold or titanium. Unstable readings might also mean poor connections, though.

Locked
Please be positive and constructive with your questions and comments.

Return to “SMT (Surface Mount Tech)”